I'm not aware of a good reason to prefer ClojureScript+Node.js in general. Maybe if you wanted interop with a specific Node library, but otherwise, Clojure on the JVM will tend to have vastly better performance, and the tooling is better.
For example, until recently ClojureScript's :keywords were just strings under the hood and it required a string compare to test equality. Apparently this is no longer the case due to this commit 2 months ago: https://github.com/clojure/clojurescript/commit/2cef52840fcd... but it shows you the kind of stuff you might run into. In a recent talk I saw by David Nolen, the lead maintainer of the ClojureScript compiler, he brought up multimethod dispatch as something that can still be as much as 100 times slower than the JVM.
As a tooling example, Clojure gives you helpful errors if you initialize records the wrong way:
user=> (defrecord Thing [foo bar])
user.Thing
user=> (Thing. 1 2) ; correct usage
#user.Thing{:foo 1, :bar 2}
user=> (Thing. 1) ; wrong number of fields
CompilerException java.lang.IllegalArgumentException: No matching ctor found for class user.Thing, compiling:(/tmp/form-init7089728177345913731.clj:1:1)
user=> (Thing 1 2) ; forgetting the dot
RuntimeException Expecting var, but Thing is mapped to class user.Thing clojure.lang.Util.runtimeException (Util.java:219)
But ClojureScript just silently gives you nils:
ClojureScript:cljs.user> (defrecord Thing [foo bar])
cljs.user/Thing
ClojureScript:cljs.user> (Thing. 1 2) ; correct usage
#cljs.user.Thing{:foo 1, :bar 2}
ClojureScript:cljs.user> (Thing. 1) ; wrong number of fields
#cljs.user.Thing{:foo 1, :bar nil}
ClojureScript:cljs.user> (Thing 1 2) ; forgetting the dot
nil